LIQUOR BISMUTHI ET AMMONITE CITRATIS. 
533 
The slight differences noticeable in the former part of the column may be 
accounted for by recollecting the great practical difficulties which attend the 
experimental determination of specific heats, and the probability that all the 
elements operated upon were not in the same physical state; some, for in¬ 
stance, being near their fusing-points, and others at a distance from them. 
The anomaly presented in the case of carbon, which, with boron and silicon 
form the only well-marked exceptions to the law, is explained by supposing 
these elements, in the condition in which they are operated upon for the de¬ 
termination of their specific heats, are in different allotropic states which do 
not permit them to exhibit the same atomic weight as they do in their 
chemical compounds. These latter follow the ordinary rules with respect to 
vapour density, the chlorides, for instance, giving numbers almost identical 
with those of theory. 
Density compared 
with H x 2. 
Molecular Weight. 
Formula. 
Chloride of Carbon (C=12) 
. . . 1564 
154 
CC1 4 
Chloride of Silicon (Si=24) 
. . . 171-5 
170 
SiCi 4 
Chloride of Boron (Bo—11) 
. . . 113-7 
1175 
BoCl 3 
Bromide of Boron .... 
... 253-6 
251 
BoBr 3 
It is, therefore, pretty certain that those atomic weights are correct which 
the chemical analogies of these bodies seem to indicate for them. 
En resume. Without attempting to explain the nature of chemical affinity, 
the ancient statement that it can be manifested only between atoms of dissi¬ 
milar properties requires modification, since we are now taught that it is 
highly probable that it may be mutually exerted by two atoms of one and the 
same element. Matter is constituted of molecules which, in all cases which 
we have been able to observe, occupy the same space when measured in the 
state of gas under uniform pressure and temperature. We have been greatly 
accustomed to the term ‘equivalent;’ this is also now restricted within 
more definite limits. An equivalent is that quantity of a substance equivalent 
to, or capable of replacing one part by weight of hydrogen. The term does 
not refer to the properties of the body, nor does the number by any means in¬ 
variably coincide with its atomic weight. The equivalent of oxygen, for ex¬ 
ample, is 8, since that is the proportion in which it combines with hydrogen, 
—16 parts to 2, or 8 to 1. Again, II 3 P0 4 represents a molecule of phos¬ 
phoric acid, not an equivalent. Strictly speaking, in the sense indicated, 
there are no such things as equivalents of saturated compounds. 
The introduction of the present system of atomic weights, and the unitary 
notation, will, it is anticipated, mark altogether a new epoch in the history of 
chemistry. Future researches will, no doubt, introduce modifications into 
this system, and may even lead to its abandonment. This good, however, 
will have been accomplished : working upon common ground, chemists will no 
longer be at issue upon points of doctrine simply from having neglected to 
agree previously upon first principles. 
(To be continued.) 
LIQUOR BISMUTHI ET AMMONLE CITRATIS. 
TO THE EDITORS OF THE PHARMACEUTICAL JOURNAL. 
Gentlemen,—I wish to make a few observations upon Mr. Schacht’s reply 
to my article on “Liquor Bismuthi et Amuioniae Citratis,” in the March 
number of your Journal. 
